Harnessing Google to Solve Parkinson's

In Wired, Thomas Goetz profiles Sergey Brin‘s search for a cure for Parkinson’s disease. Brin, who is married to genetics entrepreneur Anne Wojcicki, recently discovered he has the LRRK2 genetic mutation, which is associated with higher rates of Parkinson’s disease. What distinguishes Brin from other philanthropists, however, is his emphasis on a new model of research, known as the fourth paradigm of science: “Brin proposes a different approach, one driven by computational muscle and staggeringly large data sets. It’s a method that draws on his algorithmic sensibility-and Google’s storied faith in computing power-with the aim of accelerating the pace and increasing the potential of scientific research.”[%comments]


JS

Good luck, but I think they're over emphasizing the genetic component.

IE

Here is an idea how Google can be used to try to solve autism, or at least test one of the theories.

There is a theory that early TV exposure is one of the triggers of autism. Unfortunately, this theory is difficult to test, although there was a study showing correlation between cable subscription rates and autism rates across counties in several states.
But Google may hold the key to a better approach. Among other things they store search history for users, along with the records of what videos the users watched on YouTube.

So if we look at the users who search for autism-related items (doctors, treatment, support groups, etc.) and check whether they were actively watching toddler-age videos on YouTube 1-2 years earlier, it's possible that we may find a correlation.

This TV-trigger theory might be a long shot, but there could be other environmental factors that contribute to autism. And it could be possible to identify them by mining the Google search history database.
E.g. were the parents of autistic kids more likely to google for house paint before the birth? For new carpet? For artificial fireplace logs? No matter how ridiculous these ideas sound, at this point we are don't know what environmental factors contribute to autism. Any correlation may hold a clue to solving this $35 billion per year problem.

Read more...

JS

Solving genetic predisposition for Parkinsonism with computation power and algorithm may one day be successful. But I think it's like curing the common cold with a computer program.

Try solving real life medical issues by building a really good trivial pursuit game. Tell me where I can apply for some of that (scam)grant money.

Good luck.

Toby

The environmental factor theory of autism has pretty much been discredited.

econobiker

"What distinguishes Brin from other philanthropists, however, is his emphasis on a new model of research, known as the fourth paradigm of science: "Brin proposes a different approach, one driven by computational muscle and staggeringly large data sets. It's a method that draws on his algorithmic sensibility-"

Then how are drug companies going to charge for the research and resulting drug?

IE

Toby, as far as I know, the current consensus is that both genetic and environmental factors play some role. If it was purely genetic, then it would be impossible to explain why the disease is much more common than in the recent past (even after controlling for better detection rates).
It's the idea that vaccination is to blame was discredited, and rightfully so.

Mike

"Unfortunately, this theory is difficult to test, although there was a study showing correlation between cable subscription rates and autism rates across counties in several states."

There is also a correlation between ice cream sales in the summer and shark attacks. Does that mean they are related?